Among neutralizing antibodies (Nabs) directed against the HIV-1 envelope gp120/gp41 glycoprotein, 2F5 and 4E10 stand out for their potency and broadly neutralizing activity. Both antibodies are directed against the membrane proximal external region (MPER) of the gp41 stalk. Unfortunately, 2F5 and 4E10-like anti-MPER Nabs are rarely major components of the anti-HIV response. 2F5 and 4E10 differ from the norm in that each of their antigen binding sites contain a long, H chain complementarity determining region 3 (CDR-H3) that lacks tyrosine and includes patches of hydrophobic and charged amino acids. In previous studies we have shown that B cells bearing immunoglobulins with charged or hydrophobic CDR-H3s are normally culled from the follicular and mature, recirculating B cell subsets, i.e. the subsets from which T-dependent antibody responses are most likely to be drawn. We propose to test whether 2F5- and 4E10-like Nabs are difficult to elicit in healthy individuals because the frequency of B cells expressing Igs with this type of antigen binding site is deliberately kept low by both genetic and somatic mechanisms. We will do this by using gene targeting to replace the DH locus with a DH encoding the CDR-H3 sequence of either 2F5 or 4E10. This will force expression of a spectrum of antibodies expressing 2F5 and 4E10-like CDR-H3s. We will then test whether forced enrichment for the use of 2F5- and 4E10-derived CDR-H3 intervals will promote the production of Nabs directed against the HIV-1 MPER. To test the complementary hypothesis that the use of 2F5- and 4E10-derived CDR-H3 intervals in anti-MPER Nabs may require release from normal mechanisms of repertoire control, we will breed into these mice mutations that alter patterns of B cell signaling, repertoire selection, or cell fate outcomes. The application is responsive to the RFA in that it is specifically designed to determine "why broadly neutralizing antibodies that arise from infection are rare and how immunogens can elicit them" and to develop "more relevant animal models in which protection or enhancement of infection can be explored." Our studies also touch on "Modulating and/or specifically suppressing component aspects of the immune response to alter disease progression" in the context of determining how to manipulate the immune response in order to optimally generate protective antibodies against HIV. These studies will also serve to foster collaboration between three established basic B cell immunologists at UAB, Drs. Schroeder, Justement and Kearney;and three experts in HIV, Drs. Haynes at Duke, Grovit-Ferbas at UCLA, and Shaw at UAB. The creation of vaccines capable of eliciting neutralizing antibodies to HIV has proven more difficult than first expected because many of the antibodies that are broadly neutralizing are not only extremely uncommon;they also tend to be self-reactive. We propose to use gene targeting in mice to force B cells, including the follicular subset, to preferentially express antibody repertoires enriched for antibodies of this type, and then test whether altering BCR signaling, repertoire selection or the type of immunogen (e.g. antigen encased in liposomes or antigen displayed by pseudovirions) will facilitate the generation of antibodies that can bind the key MPER region of the HIV envelope protein in a way that will enable them to broadly neutralize HIV.